Mediaspace scheduled maintenance: Aug 25, 2026 07:00 - 12:00 AM. During this time, videos will be temporarily unavailable. Check status updates.

Lecture

Trajectory Forecasting in Autonomous Vehicles

In course

CIVIL-459: Deep learning for autonomous vehicles

Deep Learning (DL) is the subset of Machine learning reshaping the future of transportation and mobility. In this class, we will show how DL can be used to teach autonomous vehicles to detect objects, make predictions, and make decisions. (Fun fact: this s ...

Description

This lecture by the instructor covers the topic of trajectory forecasting in autonomous vehicles, focusing on deep learning models for predicting human trajectories in socially-aware transportation scenarios. The lecture delves into the challenges of sequence modelling and encoding social interactions, presenting the TrajNet++ benchmark for evaluating forecasting models. It discusses the types of trajectories, the importance of understanding complex scenarios, and the different categories of human interactions in pedestrian forecasting. The lecture also explores grid-based and non-grid-based interaction modules, attention mechanisms, and the use of transformer networks in trajectory forecasting.

Instructor

Alexandre Massoud Alahi

Alexandre Alahi is currently an Assistant Professor at EPFL. He spent five years at Stanford University as a Post-doc and Research Scientist after obtaining his Ph.D. from EPFL. His research enables machines to perceive the world and make decisions in the context of transportation problems and smart environments. He has worked on the theoretical challenges and practical applications of socially-aware Artificial Intelligence, i.e., systems equipped with perception and social intelligence. He was awarded the Swiss NSF early and advanced researcher grants for his work on predicting human social behavior. He won the CVPR Open Source Award (2012) for his work on Retina-inspired image descriptors, and the ICDSC Challenge Prize (2009) for his sparsity-driven algorithm that has tracked more than 100 million pedestrians to date. His research has been covered internationally by BBC, abc, PBS, Euronews, Wall street journal, and other national news outlets around the world. Alexandre has also co-founded multiple startups such as Visiosafe, and won several startup competitions. He was elected as one of the Top 20 Swiss Venture leaders in 2010.

Official source

Related concepts (32)

Q-learning

Q-learning is a model-free reinforcement learning algorithm to learn the value of an action in a particular state. It does not require a model of the environment (hence "model-free"), and it can handle problems with stochastic transitions and rewards without requiring adaptations. For any finite Markov decision process (FMDP), Q-learning finds an optimal policy in the sense of maximizing the expected value of the total reward over any and all successive steps, starting from the current state.

Deep learning

Deep learning is part of a broader family of machine learning methods, which is based on artificial neural networks with representation learning. The adjective "deep" in deep learning refers to the use of multiple layers in the network. Methods used can be either supervised, semi-supervised or unsupervised.

Deep reinforcement learning

Deep reinforcement learning (deep RL) is a subfield of machine learning that combines reinforcement learning (RL) and deep learning. RL considers the problem of a computational agent learning to make decisions by trial and error. Deep RL incorporates deep learning into the solution, allowing agents to make decisions from unstructured input data without manual engineering of the state space. Deep RL algorithms are able to take in very large inputs (e.g.

Reinforcement learning

Reinforcement learning (RL) is an area of machine learning concerned with how intelligent agents ought to take actions in an environment in order to maximize the notion of cumulative reward. Reinforcement learning is one of three basic machine learning paradigms, alongside supervised learning and unsupervised learning. Reinforcement learning differs from supervised learning in not needing labelled input/output pairs to be presented, and in not needing sub-optimal actions to be explicitly corrected.

Autonomous robot

An autonomous robot is a robot that acts without recourse to human control. The first autonomous robots environment were known as Elmer and Elsie, which were constructed in the late 1940s by W. Grey Walter. They were the first robots in history that were programmed to "think" the way biological brains do and meant to have free will. Elmer and Elsie were often labeled as tortoises because of how they were shaped and the manner in which they moved. They were capable of phototaxis which is the movement that occurs in response to light stimulus.

Related lectures (37)

Perception: Data-Driven Approaches

Explores perception in deep learning for autonomous vehicles, covering image classification, optimization methods, and the role of representation in machine learning.

Autonomous Vehicles: Trajectory Prediction and Social Behavior

Explores challenges in deep learning for autonomous vehicles, emphasizing social behavior modeling and feasible trajectory prediction.

Neural Networks Optimization

Explores neural networks optimization, including backpropagation, batch normalization, weight initialization, and hyperparameter search strategies.

Deep Learning for Autonomous Vehicles: Predictive Models

Explores predictive models and trackers for autonomous vehicles, covering object detection, tracking challenges, neural network-based tracking, and 3D pedestrian localization.

Sequence to Sequence Models: Overview and Applications